In a CDMA communication system, a data transmission synchronization problem is very important, if the data transmission synchronization problem of a receiving end and a sending end, it may be caused that a transmission device cannot perform normal communication. There are many traditional CDMA system synchronization methods, for example, in a synchronization capture phase, the sending end first sends a synchronization sequence to the receiving end, and the receiving end establishes synchronization between the receiving end and the sending end after receiving the synchronization sequence of the sending end, and then switches to a data state to perform data transmission. This synchronization method adopts different Pseudo-Noise (PN) codes in a synchronization signal capture phase and a data processing phase. Currently, a typical representative synchronization algorithm of a spreading chip STEL-2000A adopts twice sampling at a front end of a digital matched filter, a sample value after twice sampling and weighted averaging is sent to the matched filter.
A similarity of a traditional synchronization method is to use a relevant feature of the PN code to perform synchronization of the receiving end and the sending end, a difference is a chip sampling time of the front end of the matched filter. However, with regard to a relatively maturely applied currently CDMA synchronization technique, a wrong sampling situation at the front end of the matched filter is neglected. Specifically, an ideal sampling situation is shown in FIG. 1 (a), that is, chips di−1, di, di+1 sample twice separately, and samples correspond to each chip separately are ai−1, bi−1, ai, bi, ai+1, bi+1, and weighted average is performed on information sampled by each chip, namely (ai−1+bi−1)/2, (ai+bi)/2, (ai+1+bi+1)/2. However, in actual operation, this situation may exist: the weighted twice samples do not come from the same chip, but come from adjacent chips, as shown in FIG. 1 (b) or FIG. 2, after the sample is weighted, (ai+bi)/2 is sent to the matched filter, because ai and bi do not come from the same chip, that is wrong sampling is caused. In addition, although it is in the ideal sampling situation, wrong sampling may be caused all because of data jitter, delay, jitter of a local clock, and a Doppler effect. The wrong sampling may bring a serious inter-chip interference, the wrong sampling at a place at which a positive electrical level and a negative electrical level of a PN chip alternate results in a very small weighted sample value, and finally results in a high bit error rate of the receiving end.
Aiming at a problem in the related art that a CDMA synchronization method possesses the wrong sampling situation, which results in a high bit error rate of the receiving end, an effective solution has not been proposed currently.